Trent AI, a London-based provider of context-driven security for agentic AI applications, has raised $13M in seed funding led by LocalGlobe and Cambridge Innovation Capital. The platform deploys specialized AI agents in a continuous Scan-Judge-Mitigate-Evaluate loop to secure agents, code, infrastructure, and workflows. The capital will fuel product expansion and team growth following emergence from stealth.
Agentic Security Funding Surges
The round arrives amid surging investment in AI agent security. Oasis Security raised $120M in March 2026 for AI agent identity tools, while Above Security secured $50M to address insider threats from agents. Trent AI's compounding intelligence loop targets dynamic risks in dev workflows like Claude Code and Lovable that identity-focused rivals overlook.
Governance Lags Agent Adoption
A Deloitte survey finds 74% of companies plan agentic AI deployment within two years, yet only 21% have mature governance. Traditional security tools generate alert noise without context, missing agent-specific threats like tool misuse or data exfiltration. Agentic systems introduce autonomous behaviors that static scanners cannot assess.
Multi-Agent Loop Secures Workflows
Trent AI builds security-by-design with AI agents that scan for vulnerabilities, judge risks contextually using reinforcement learning, mitigate automatically, and evaluate outcomes to improve over time. Integrations with Lovable, Claude Code, and OpenClaw enable seamless workflow security without manual triage. This self-reinforcing approach contrasts with model-scanning competitors like Protect AI.
As Eno Thereska, Co-Founder and CEO, noted:
“Organisations are deploying AI agents and autonomous workflows faster than their security can adapt… Trent AI is tackling these difficult and important problems.”
Cambridge Roots Drive Differentiation
Early design partners including Canopy, Commscentre, and ML@CMU validate the platform's fit for agent builders. The loop's judgment layer addresses AI's core issue: not knowledge, but contextual reasoning for threats.
UK VCs Bet on Deep Tech
LocalGlobe, with unicorns like Mistral AI and Monzo, brings seed-stage expertise in AI and security, including recent bet on Opti IAM. Cambridge Innovation Capital, tied to the university ecosystem, backs deep tech like Riverlane quantum. Angels from OpenAI, Spotify, AWS, and Databricks add applied AI credibility, signaling conviction in agentic security's emergence.
Saul Klein of LocalGlobe said:
“The rise of agents goes hand in hand with the rise of new security threats… Trent AI is uniquely positioned.”
AI Cybersecurity Market Doubles
The AI cybersecurity market stands at $25.53B in 2026, projected to reach $50.83B by 2031 at 14.9% CAGR. Agentic AI alone hits $9.14B in 2026, growing to $139B by 2034. Enterprises shift from copilots to production agents, amplifying demand for runtime protection beyond prompt defenses like Lakera.
FAANG Vets Lead Agent Defense
Co-founders bring elite pedigrees: CEO Eno Thereska led AI data protection at Alcion, acquired by Veeam; Chief Scientist Neil Lawrence holds the DeepMind Chair at Cambridge; CTO Zhenwen Dai headed Spotify's RL playlist lab. This mix of cloud infra, ML research, and production AI scales secure agentic systems. All have AWS and big tech experience, ensuring practical deployment.
Platform Integrations Expand Rapidly
Recent launches include Security Advisor for Lovable and OpenClaw skill, with OWASP partnership and RSAC 2026 presence. Design partners report faster vulnerability ID and remediation, positioning Trent AI for national scaling via agent ecosystems.